Photolyase Mechanism Can Proceed through an Oxetane Intermediate

Sep 3, 2008 - UniVersité de Reims Champagne Ardenne, Institut de Chimie Moléculaire de Reims, CNRS UMR 6229, UFR de. Pharmacie, 51 rue ...
0 downloads 0 Views 93KB Size
Published on Web 09/03/2008

Evidence That the (6-4) Photolyase Mechanism Can Proceed through an Oxetane Intermediate Saaˆdia Asgatay,† Christian Petermann,† Dominique Harakat,‡ Dominique Guillaume,† John-Stephen Taylor,§ and Pascale Clivio†,* UniVersite´ de Reims Champagne Ardenne, Institut de Chimie Mole´culaire de Reims, CNRS UMR 6229, UFR de Pharmacie, 51 rue Cognacq-Jay, 51096 Reims Cedex, France, UFR des Sciences Exactes et Naturelles, Baˆtiment 18, Europol’Agro, BP 1039, 51687 Reims cedex 2, France, and Department of Chemistry, Washington UniVersity, St. Louis, Missouri 63130 Received July 7, 2008; E-mail: [email protected]

Pyrimidine (6-4) pyrimidone photoproducts ((6-4) PPs) are one of the main UV photoproducts formed in DNA under solar UV light.1 This DNA photodamage, produced at two adjacent pyrimidines, is implicated in the etiology of skin cancer and photoageing.2 Elucidation of the striking chemical structure of (6-4) PPs in 1967 simultaneously raised the problem of their formation mechanism. It is now accepted that (6-4) PPs result from the isomerization, through ring opening triggered by the deprotonation of N3 of the 3′-pyrimidine, of an instable four-membered heterocycle intermediate initially produced by a Paterno`-Bu¨chi reaction between the C5-C6 double bond of a 5′-pyrimidine and the C4-heteroatom of a 3′-pyrimidine.3 Whereas pieces of evidence for an unstable oxetane intermediate were reported in 1969,4 the only experimental proof supporting this intricate mechanism was brought in the 1990s through the characterization of stable thietanes using the photochemistry of 4-thiothymine derivatives.5 Recently, the discovery of a photolyase specific for the repair of (6-4) PPs has renewed the interest in the oxetane which is also supposed to be the key intermediate in the repair mechanism.6 Oxetane formation from (6-4) PP is believed to be an acido/basic catalytic process triggered by the protonation of the N3 atom of the pyrimidone moiety and deprotonation of the 5-OH of the 5-hydroxy-5,6-dihydrothymine residue by two histidine residues of the (6-4) photolyase (Scheme 1).6d,e Surprisingly, despite the considerable interest currently devoted to the formation and repair mechanism of (6-4) PPs, neither successful experimental attempts at tackling the oxetane intermediate reactivity nor at synthesizing oxetane derivatives close to the biologically relevant one have ever been reported.7 Herein, we provide new insights on formation of the putative oxetane and provide experimental evidence that complement the current model for the repair mechanism of (6-4) PPs by (6-4) photolyase. Guided by our photochemical studies in the 4-thiothymine series,5b,8 we decided to investigate the photoreactivity of Tpm3T (1)9 at 254 nm anticipating that substitution of the N3 atom of the 3′-end thymine residue should either stop the photochemical reaction at the oxetane level (4) (Scheme 2) or lead to the N3-methyl iminium derivative (5) (Scheme 2, path a), a close analogue of the proposed iminium intermediate in the repair process (Scheme 1). UV-C irradiation of 1 in aqueous solution (pH 7, 10 mM Na phosphate) gave rise to 2 and 3 (Supporting Information Figures S4 and S9) isolated after HPLC in 15% and 22% yield, respectively, together with 57% yield of recovered 1. Photoproduct 3 was identified as the cis-syn N3-methyl cyclobutane photoproduct from its UV, mass † ‡ §

UMR 6229, UFR de Pharmacie. UMR 6229, UFR des Sciences Exactes et Naturelles. Washington University.

12618

9

J. AM. CHEM. SOC. 2008, 130, 12618–12619

Scheme 1

Scheme 2

(HRMS (ESI, (M + Na)+): calcd for C21H29N4O12PNa, 583.1417; found, 583.1423), and NMR data (S15-17), and comparison with literature.10 For photoproduct 2, absence of UV absorption maximum above 240 nm signed the loss of the two thymine chromophores. NMR data of the Tp- base moiety of 2 (δ H6 4.90, δ C6 58.7; δH CH3 1.66, δC CH3 28.4; δ C5 71.0)11 were very close to those of the Tp- part of the (6-4) or Dewar of TpT.10,12 However, NMR data of the -pm3T base moiety of 211 particularly δ C6 77.6 and C4 131.4 were not consistent with the oxetane structure 4. Interestingly, δ C6 and δ C4 were in accordance with the presence of an hydroxyl group at the C6 position and a C4-C5 insaturation (δ C5 123.5 and δ C4 131.4). Therefore, we identified 2 as the N3-methyl-C6 hydrate analogue of the (6-4) photoproduct of TpT. The ESI mass spectrum of 2 recorded in a methanol/H2O mixture (negative mode, SI 25) afforded the expected ion at m/z 577 (M - H)- (calcd for C21H30N4O13P, 577.1547; found, 577.1560). The presence of an ion at m/z 559 indicated that this product was readily dehydrated (calcd for C21H28N4O12P, 559.1441; found, 559.1440). Fast exchange of the OH at C6 with a OCH3, ascertained by the presence of an ion at m/z 591, (calcd for C22H32N4O13P, 591.1704; found, 591.1687) fully supported the expected chemical reactivity of 2. MS/MS performed on the ions 10.1021/ja805214s CCC: $40.75  2008 American Chemical Society

COMMUNICATIONS

at m/z 591, 577, and 559 (SI 26) yielded a sole major ion at m/z 446 representing the loss of MeOH-H2O-113, H2O-113, and 113, respectively. Loss of a 113 neutral fragment is characteristic of the Tp- part of the (6-4) adduct of TpT.13 NOEs between H6 and H3′ in each residue and between H6 Tp- and both CH3 Tp- and C5-CH3 -pm3T indicated that the photochemical reaction arose between the two nucleoside residues in the anti glycosyl bond conformation and a C6 S configuration of the -pm3T residue. Formation of 2 is fully consistent with the initial formation of an oxetane intermediate. From our observations, it is likely that oxetane 4 is so unstable that it leads to 2 either by addition of H2O at C6 (path b, scheme 2) or via the 1,2-dihydro-2-oxopyrimidinium cation 5 (path a, Scheme 2), which is in equilibrium14 with its pseudobase 2. The diastereospecific formation of the C6 S isomer is in accordance with water addition on the less hindered side of the molecule. Then, we anticipated that if a minute fraction of 2 was present as its 2-oxopyrimidinium cationic form (5), intramolecular Tp- C5 hydroxyl addition at position C4 of -pm3T would lead to the oxetane that would reverse to 1 under 254 nm light (Scheme 2). Therefore, we exposed 2 to UV-C. To our delight, this led to its reversal to the parent dinucleotide 1 (S5).15 Upon standing in H2O at neutral pH, 2 gives rise to 6. This derivative is directly obtained, together with 3, when 1 is exposed to the 254 nm light in unbuffered water solution (pH 4.6) (S4, S18). In such conditions, 2 is not detected. The 1H NMR spectrum of 6 displayed two singlets integrating for one proton each at δ 8.40 and 5.51 and two C5 methyl groups at δ 2.09 and 1.76 reminiscent of a (6-4) structure as also indicated by its UV absorption maximum at 321 nm. The molecular weight of 6 (HRMS (ESI, (M - H)-): calcd for C21H30N4O13P, 577.1547; found, 577.1532) was 18 amu higher than that of 1. Whereas the NMR data of the Tp-moiety of 6 were close to the corresponding part of 2, hydrolysis of the 3′-end glycosyl bond of 6 was ascertained by the presence of split groups of signals for this sugar residue in 1H and 13C NMR spectra together with the typical chemical shift of its C1′ (δ 99.5).16 The HMQC spectrum of 6 displayed a correlation between the proton at δ 8.50 attributed to H6 of the -pm3T part and its carbon at δ 170.8. This H6 -pm3T proton coupled with two quaternary carbons at δ 116.0 and δ 155.2, attributed respectively to C5 and C4. The chemical shifts of the -pm3T base moiety are very close to the corresponding ones of the known 5, N3-dimethyl-2-pyrimidinone motif.8b Therefore, 6 derives from the hydrolysis of the 3′-end N-glycosyl bond of 2 and aromatization, through dehydration of the -pm3T motif and/or the direct hydrolysis of the 3′-end N-glycosyl bond of 5 that is in equilibrium with 2 (Scheme 2). In conclusion, our results are of utmost importance with regard to the repair mechanism of (6-4) PPs by (6-4) photolyase. Indeed, we provide the first experimental evidence showing that as suggested, an iminium ion is able to restore the oxetane intermediate. Our study strongly reinforces the proposed model for (6-4) photoenzymatic repair and highlights a possible, and yet unsuspected, reactivity for the iminium ion intermediate that can readily

react with water addition at the C6 position or at position C1′ of the 3′-end sugar residue. Spontaneous 3′-end N-glycosyl bond rupture is another outcome that may compete with oxetane formation within cellular DNA. This scenario would generate an abasic site which cannot be repaired by the (6-4) photolyase. Acknowledgment. CNRS and Region Champagne Ardenne are gratefully acknowledged for a postdoctoral fellowship to S.A. and financial support, respectively. We thank A. Martinez for recording some NMR spectra. Part of this work was supported by a IARCWHO fellowship to P.C. Supporting Information Available: Experimental conditions for UV irradiation and HPLC, chromatograms of the photolysis of 1, and the 254 nm photoreversion of 2 and 3. NMR spectra of 1, 2, 3, 6, of the 254 nm photolysis of 1, ESI MS data for 2. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Cadet, J.; Courdavault, S.; Ravanat, J.-L.; Douki, T. Pure Appl. Chem. 2005, 77, 947–961. (2) (a) Taylor, J.-S. Pure Appl. Chem. 1995, 67, 183–190. (b) Matsumura, Y.; Ananthaswamy, H. N. Front. Biosci. 2002, 7, d765-783. (c) Trautinger, F. Clin. Dermatol. 2001, 26, 573–577. (3) (a) Wang, S. Y.; Varghese, A. J. Biochem. Biophys. Res. Commun. 1967, 29, 543–549. (b) Varghese, A. J.; Wang, S. Y. Science 1968, 160, 186– 188. (4) Rahn, R. O.; Hosszu, J. L. Photochem. Photobiol. 1969, 10, 131–137. (5) (a) Clivio, P.; Fourrey, J.-L.; Gasche, J.; Favre, A. J. Am. Chem. Soc. 1991, 113, 5481–5483. (b) Liu, J.; Taylor, J.-S. J. Am. Chem. Soc. 1996, 118, 3287–3288. (6) (a) Todo, T.; Takemori, H.; Ryo, H.; Ihara, M.; Matsunaga, T.; Nikaido, O.; Sato, K.; Nomura, T. Nature 1993, 361, 371–374. (b) Kim, S,-T.; Malhotra, K.; Smith, C. A.; Taylor, J.-S.; Sancar, A. J. Biol. Chem. 1994, 269, 8535–8540. (c) Zhao, X.; Liu, J.; Hsu, D. S.; Zhao, S.; Taylor, J.-S.; Sancar, A. J. Biol. Chem. 1997, 272, 32580–32590. (d) Hitomi, K.; Nakamura, H.; Kim, S.-T.; Mizukoshi, T.; Ishikawa, T.; Iwai, S.; Todo, T. Biol. Chem. 2001, 276, 10103–10109. (e) Schleicher, E.; Hitomi, K.; Kay, C. W. M.; Getzoff, E. D.; Todo, T.; Weber, S. J. Biol. Chem. 2007, 282, 4738–4747 For selected reviews, see: (f) Sancar, A. Chem. ReV. 2003, 103, 2203–2237. (g) Weber, S. Biochim. Biophys. Acta 2005, 1707, 1–23. (7) Simple oxetane models have been prepared for artificial repair studies: Selected examples: (a) Joseph, A.; Prakash, G.; Falvey, D. E. J. Am. Chem. Soc. 2000, 122, 11219–11225. (b) Cichon, M. K.; Arnold, S.; Carell, T. Angew. Chem., Int. Ed. 2002, 41, 767–770. (c) Boussicault, F.; Robert, M. J. Phys. Chem. B 2006, 110, 21987–21993. (d) Song, Q.-H.; Wang, H.-B.; Tang, W.-J.; Guo, Q.-X.; Yu, S.-Q. Org. Biomol. Chem. 2006, 4, 291– 298. (e) Trzcionka, J.; Lhiaubet-Vallet, V.; Paris, C.; Belmadoui, N.; Climent, M. J.; Miranda, M. A. ChemBioChem. 2007, 8, 402–407. (8) (a) Clivio, P.; Fourrey, J.-L.; Gasche, J.; Favre, A. Tetrahedron Lett. 1992, 33, 1615–1618. (b) Clivio, P.; Fourrey, J.-L.; Szabo, T.; Stawinski, J. J. Org. Chem. 1994, 59, 7273–7283. (9) The synthesis of 1 (HRMS (ESI, (M + H)+): calcd for C21H30N4O12P, 561.1598; found, 561.1599) will be reported later. See S78 for 1H and 13C NMR spectra, respectively. (10) Rycyna, R. E.; Alderfer, J. Nucleic Acids Res. 1985, 13, 5949–5963. (11) Attribution of Tp- and -m3pT sugar signals was made from the multiplicity of H3′ Tp- signal and COSY; attribution of Tp- and -pm3T aglycone signals was made from LR correlations between H6-C1′; H6-CH3; H6-C2; H6-C5 and CH3-C4 for each residue. (12) Taylor, J.-S.; Garrett, D.; Cohrs, M. P. Biochemistry 1988, 27, 7206–7215. (13) Wang, Y.; Taylor, J.-S.; Gross, M. L. Chem. Res. Toxicol. 1999, 12, 1077– 1082. (14) Tee, O. S.; Endo, M. Can. J. Chem. 1976, 54, 2681–2688. (15) Exposure of 3 to 254 nm also gave rise to 1 as expected for a cyclobutane photoproduct (S6). (16) Breitmaier, E.; Voelter, W. Carbon-13 NMR Spectroscopy; Ebel, H. F., Ed.; VCH: Weinheim, Germany, 1987.

JA805214S

J. AM. CHEM. SOC.

9

VOL. 130, NO. 38, 2008

12619